Yes.Does it include fix for wireless mentioned in some thread for test ROSrc12 that wasn't officially released?
More importantly, is it working 100%?
v5.0 has those fixes as well.Does this include the fix for the RB1100 ethernet port issues like 4.17?
If you see my post above I think it is related to the previous install having the NV2 test package installed. Hopefully they can fix this quickly.one time i updated router trough wireless.
wireless package seems to not be saved or damaged during upload to the router
be in mind to check the list of uploaded packages before "/system reboot"
What version did you upgrade from? 4.X or 5.RC?hm... i do the same, no problem on my router
Mode station-bridge
This mode works only with RouterOS APs and provides support for transparent protocol-independent L2 bridging on station device. RouterOS AP accepts clients in station-bridge mode when enabled using bridge-mode parameter. In this mode AP maintains forwarding table with information on what MAC addresses are reachable over which station device.
This mode is MikroTik proprietary and can't be used to connect other brand devices.
This mode is safe to use for L2 bridging and should be used whenever there are sufficient reasons to not use station-wds mode.
So is it the version of user manager that we've all been commenting about in this thread?There is no old user manager at v5.0.
???wirelessrudy, that's why you have to test rc11 to see if your requests have been made or not.
sergejs, I am not referring to of one of the ´Mode´ options in the top of the winbox Wireless / Interface / Wireless-tab window.Wirelessrudy, about bridge mode from wiki,
http://wiki.mikrotik.com/wiki/Manual:Wi ... ion-bridge
Mode station-bridge
This mode works only with RouterOS APs and provides support for transparent protocol-independent L2 bridging on station device. RouterOS AP accepts clients in station-bridge mode when enabled using bridge-mode parameter. In this mode AP maintains forwarding table with information on what MAC addresses are reachable over which station device.
This mode is MikroTik proprietary and can't be used to connect other brand devices.
This mode is safe to use for L2 bridging and should be used whenever there are sufficient reasons to not use station-wds mode.
0 device="03:00.0" name="unknown device (rev: 3)" vendor="NEC Corporation"
category="USB Controller" vendor-id="0x1033" device-id="0x0194" irq=10
memory=0xF7D00000-0xF7DFFFFF
1 device="01:00.1" name="82576 Gigabit Network Connection (rev: 1)"
vendor="Intel Corporation" category="Ethernet controller"
vendor-id="0x8086" device-id="0x10c9" irq=10
memory=0xF7C00000-0xF7C1FFFF,0xF7000000-0xF73FFFFF,0xF7C40000-0xF7C43FFF
io=0xE000-0xEFFF
2 device="01:00.0" name="82576 Gigabit Network Connection (rev: 1)"
vendor="Intel Corporation" category="Ethernet controller"
vendor-id="0x8086" device-id="0x10c9" irq=11
memory=0xF7C20000-0xF7C3FFFF,0xF7800000-0xF7BFFFFF,0xF7C44000-0xF7C47FFF
io=0xE020-0xE03F
3 device="00:1f.3" name="Cougar Point SMBus Controller (rev: 4)"
vendor="Intel Corporation" category="SMBus" vendor-id="0x8086"
device-id="0x1c22" irq=11 memory=0xF7E24000-0xF7E240FF io=0xF040-0xF05F
4 device="00:1f.2" name="Cougar Point 4 port SATA IDE Controller (rev: 4)"
vendor="Intel Corporation" category="IDE interface" vendor-id="0x8086"
device-id="0x1c00" irq=4 io=0xF090-0xF09F,0xF080-0xF08F
5 device="00:1f.0" name="unknown device (rev: 4)" vendor="Intel Corporation"
category="ISA bridge" vendor-id="0x8086" device-id="0x1c4a" irq=0
6 device="00:1d.0"
name="Cougar Point USB Enhanced Host Controller #1 (rev: 4)"
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x1c26" irq=7 memory=0xF7E25000-0xF7E253FF
7 device="00:1c.1" name="Cougar Point PCI Express Root Port 2 (rev: 180)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x1c12" irq=11
8 device="00:1c.0" name="Cougar Point PCI Express Root Port 1 (rev: 180)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x1c10" irq=10
9 device="00:1b.0"
name="Cougar Point High Definition Audio Controller (rev: 4)"
vendor="Intel Corporation" category="Audio device" vendor-id="0x8086"
device-id="0x1c20" irq=3 memory=0xF7E20000-0xF7E23FFF
10 device="00:1a.0"
name="Cougar Point USB Enhanced Host Controller #2 (rev: 4)"
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x1c2d" irq=11 memory=0xF7E26000-0xF7E263FF
11 device="00:19.0" name="unknown device (rev: 4)" vendor="Intel Corporation"
category="Ethernet controller" vendor-id="0x8086" device-id="0x1503" irq=5
memory=0xF7E00000-0xF7E1FFFF,0xF7E27000-0xF7E27FFF io=0xF060-0xF07F
12 device="00:16.0" name="Cougar Point HECI Controller #1 (rev: 4)"
vendor="Intel Corporation" category="Communication controller"
vendor-id="0x8086" device-id="0x1c3a" irq=11 memory=0xF7E28000-0xF7E2800F
13 device="00:02.0" name="Sandy Bridge Integrated Graphics Controller (rev: 9)>
vendor="Intel Corporation" category="VGA compatible controller"
vendor-id="0x8086" device-id="0x0102" irq=11
memory=0xF6800000-0xF6BFFFFF,0xE0000000-0xEFFFFFFF io=0xF000-0xF03F
14 device="00:01.0" name="unknown device (rev: 9)" vendor="Intel Corporation"
category="PCI bridge" vendor-id="0x8086" device-id="0x0101" irq=11
15 device="00:00.0" name="unknown device (rev: 9)" vendor="Intel Corporation"
category="Host bridge" vendor-id="0x8086" device-id="0x0100" irq=0
0 device="04:00.0" name="RTL-8139/8139C/8139C+ (rev: 16)"
vendor="Realtek Semiconductor Co., Ltd." category="Ethernet controller"
vendor-id="0x10ec" device-id="0x8139" irq=5 memory=0xFD9FF000-0xFD9FF0FF
io=0xDE00-0xDEFF
1 device="03:00.0" name="unknown device (rev: 192)"
vendor="Attansic Technology Corp." category="Ethernet controller"
vendor-id="0x1969" device-id="0x1063" irq=12 memory=0xFDCC0000-0xFDCFFFFF
io=0xBF00-0xBF7F
2 device="01:00.1" name="82571EB Gigabit Ethernet Controller (rev: 6)"
vendor="Intel Corporation" category="Ethernet controller"
vendor-id="0x8086" device-id="0x105e" irq=12
memory=0xFD7A0000-0xFD7BFFFF,0xFD780000-0xFD79FFFF io=0xAE00-0xAE1F
3 device="01:00.0" name="82571EB Gigabit Ethernet Controller (rev: 6)"
vendor="Intel Corporation" category="Ethernet controller"
vendor-id="0x8086" device-id="0x105e" irq=10
memory=0xFD7E0000-0xFD7FFFFF,0xFD7C0000-0xFD7DFFFF io=0xAF00-0xAF1F
4 device="00:1f.3" name="82801G (ICH7 Family) SMBus Controller (rev: 1)"
vendor="Intel Corporation" category="SMBus" vendor-id="0x8086"
device-id="0x27da" irq=11 io=0x500-0x51F
5 device="00:1f.2"
name="82801GB/GR/GH (ICH7 Family) SATA IDE Controller (rev: 1)"
vendor="Intel Corporation" category="IDE interface" vendor-id="0x8086"
device-id="0x27c0" irq=11 io=0xF300-0xF30F
6 device="00:1f.1" name="82801G (ICH7 Family) IDE Controller (rev: 1)"
vendor="Intel Corporation" category="IDE interface" vendor-id="0x8086"
device-id="0x27df" irq=255 io=0xF800-0xF80F
7 device="00:1f.0"
name="82801GB/GR (ICH7 Family) LPC Interface Bridge (rev: 1)"
vendor="Intel Corporation" category="ISA bridge" vendor-id="0x8086"
device-id="0x27b8" irq=0
8 device="00:1e.0" name="82801 PCI Bridge (rev: 225)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x244e" irq=0
9 device="00:1d.7" name="82801G (ICH7 Family) USB2 EHCI Controller (rev: 1)"
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x27cc" irq=3 memory=0xFDFFF000-0xFDFFF3FF
10 device="00:1d.3" name="82801G (ICH7 Family) USB UHCI Controller #4 (rev: 1)>
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x27cb" irq=10 io=0xFB00-0xFB1F
11 device="00:1d.2" name="82801G (ICH7 Family) USB UHCI Controller #3 (rev: 1)>
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x27ca" irq=9 io=0xFC00-0xFC1F
12 device="00:1d.1" name="82801G (ICH7 Family) USB UHCI Controller #2 (rev: 1)>
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x27c9" irq=11 io=0xFD00-0xFD1F
13 device="00:1d.0" name="82801G (ICH7 Family) USB UHCI Controller #1 (rev: 1)>
vendor="Intel Corporation" category="USB Controller" vendor-id="0x8086"
device-id="0x27c8" irq=3 io=0xFE00-0xFE1F
14 device="00:1c.1" name="82801G (ICH7 Family) PCI Express Port 2 (rev: 1)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x27d2" irq=12
15 device="00:1c.0" name="82801G (ICH7 Family) PCI Express Port 1 (rev: 1)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x27d0" irq=10
16 device="00:02.0"
name="82G33/G31 Express Integrated Graphics Controller (rev: 16)"
vendor="Intel Corporation" category="VGA compatible controller"
vendor-id="0x8086" device-id="0x29c2" irq=10
memory=0xFDF00000-0xFDF7FFFF,0xD0000000-0xDFFFFFFF,0xFD600000-0xFD6FFFFF
io=0xFF00-0xFF07
17 device="00:01.0"
name="82G33/G31/P35/P31 Express PCI Express Root Port (rev: 16)"
vendor="Intel Corporation" category="PCI bridge" vendor-id="0x8086"
device-id="0x29c1" irq=10
18 device="00:00.0" name="82G33/G31/P35/P31 Express DRAM Controller (rev: 16)"
vendor="Intel Corporation" category="Host bridge" vendor-id="0x8086"
device-id="0x29c0" irq=0
No chance. RB133c died while making supout.rif.ste, we are waiting for support output file from the router.
Same problem. RB433 4.16 to 5.0 Upgrade Fail (Disconnect)I upgraded an RB running 4.16 to 5.0 that had the NV2 package installed and when the router came back up it didn't have any wireless package enabled and lost all the wireless settings. Is there any way around this issue?
I lucked out in this case because it was the local side that I upgraded, but it could have resulted in a frustrating truckroll.
Thank you for info. But the second box is not so new and it also has unknown pci devices... Can you help?I suppose you mean Intel Ethernet controller is not supported on your Sandy Bridge motherboard.
Currently there is no drivers for it.
at first logon feature is removed from User Manager.
Currently there is no plans to get it back in the near future.
Please read this section below again and look at the sentence in bold, it will help you to understand that that feature allows or disallows AP to accept station-bridge mode clients.sergejs, I am not referring to of one of the ´Mode´ options in the top of the winbox Wireless / Interface / Wireless-tab window.
I am referring to the new "Bridge Mode" with option "enaled / disabled" only under the "WMM Support" field in that same winbox subwindow.
This is new in ros5.0. If it is the same as the option you can make at the top in that window than why a second button to enable/disable that option again?
Makes not a lot of sense IMHO.
Mode station-bridge
This mode works only with RouterOS APs and provides support for transparent protocol-independent L2 bridging on station device. RouterOS AP accepts clients in station-bridge mode when enabled using bridge-mode parameter. In this mode AP maintains forwarding table with information on what MAC addresses are reachable over which station device.
This mode is MikroTik proprietary and can't be used to connect other brand devices.
This mode is safe to use for L2 bridging and should be used whenever there are sufficient reasons to not use station-wds mode.
I was just realizing what these words were meaning when this post come at my notice. I understand and bow my head now.... Thanks anyway for the explanation. Probably I am not the only one not reading thorough or understanding all well. So now we have extra explanation.Please read this section below again and look at the sentence in bold, it will help you to understand that that feature allows or disallows AP to accept station-bridge mode clients.sergejs, I am not referring to of one of the ´Mode´ options in the top of the winbox Wireless / Interface / Wireless-tab window.
I am referring to the new "Bridge Mode" with option "enaled / disabled" only under the "WMM Support" field in that same winbox subwindow.
This is new in ros5.0. If it is the same as the option you can make at the top in that window than why a second button to enable/disable that option again?
Makes not a lot of sense IMHO.Mode station-bridge
This mode works only with RouterOS APs and provides support for transparent protocol-independent L2 bridging on station device. RouterOS AP accepts clients in station-bridge mode when enabled using bridge-mode parameter. In this mode AP maintains forwarding table with information on what MAC addresses are reachable over which station device.
This mode is MikroTik proprietary and can't be used to connect other brand devices.
This mode is safe to use for L2 bridging and should be used whenever there are sufficient reasons to not use station-wds mode.
These links are still running fine mids of the day with plenty of traffic etc. One by one and little by little I am upgrading now some remote clients (mostly 133C's and 411's) from different AP's and so far no disaster. Fingers crossed!Updated two rb433's forming a remote link;
packages: advanced-tools, dhcp, routerboard, routing, system and wireless.
upgraded firmware.
So far so good. Link is running in 5Ghz 10Mhz channel Management protection enabled and 802.11 protocol with set data rates.
Switched link to run with nv2 protocol; so far so good. No probs.
Second unit is also AP, AP running in 2,4Ghz normal 802.11 + WPA PSK/WPA2PSK: No issue so far.
Link is just upgrade in the middle of the night and adjacent network don't see effect of nv2.
Lets wait what the day will bring... fingers crossed..
please make the support output file and contact support@mikrotik.comI have problem with CPU load 100% after upgrading from 4.16 to 5.0 all the time. Before it was never 100% all the time.
I can't upgrade firmware from 2.28 to 2.33. I give upgrade command and get message that it is upgraded sucessfully, and to reboot it. After reboot I see that it is still on 2.28 version firmware, and that available version for upgrade is 2.33.
Also, in winbox new version when I open interfaces windows and ppp window I don't see anything. After some time I get message in this window ERROR: timeout
It is RB750 router. Can anyone help me with this problems.
Regards,
Sasa
... Well I really hope this is an April fools joke...you are wrong, Nv2 is very much architecture dependent, there are many technical reasons which cause difficulties.
Not a joke, it's been noted for at least a few days that it doesn't work with x86... although if you look back, at least 1 person got nv2 to work with x86 anyway.... Well I really hope this is an April fools joke...you are wrong, Nv2 is very much architecture dependent, there are many technical reasons which cause difficulties.
Oh my soul!Not a joke, it's been noted for at least a few days that it doesn't work with x86... although if you look back, at least 1 person got nv2 to work with x86 anyway.... Well I really hope this is an April fools joke...you are wrong, Nv2 is very much architecture dependent, there are many technical reasons which cause difficulties.
The same problem but fw upgrade passed successfully from terminal window.I have problem with CPU load 100% after upgrading from 4.16 to 5.0 all the time. Before it was never 100% all the time.
I can't upgrade firmware from 2.28 to 2.33. I give upgrade command and get message that it is upgraded sucessfully, and to reboot it. After reboot I see that it is still on 2.28 version firmware, and that available version for upgrade is 2.33.
Also, in winbox new version when I open interfaces windows and ppp window I don't see anything. After some time I get message in this window ERROR: timeout
It is RB750 router. Can anyone help me with this problems.
Regards,
Sasa
Oh my soul!Not a joke, it's been noted for at least a few days that it doesn't work with x86... although if you look back, at least 1 person got nv2 to work with x86 anyway.
... Well I really hope this is an April fools joke...
And this just as I was about to install 5.0 onto an x86 machine.
That was a very near miss.
The same problem but fw upgrade passed successfully from terminal window.I have problem with CPU load 100% after upgrading from 4.16 to 5.0 all the time. Before it was never 100% all the time.
I can't upgrade firmware from 2.28 to 2.33. I give upgrade command and get message that it is upgraded sucessfully, and to reboot it. After reboot I see that it is still on 2.28 version firmware, and that available version for upgrade is 2.33.
Also, in winbox new version when I open interfaces windows and ppp window I don't see anything. After some time I get message in this window ERROR: timeout
It is RB750 router. Can anyone help me with this problems.
Regards,
Sasa
I am a trusting person. I read the changelog and there was nothing there that I saw.How was that a near miss??? It is very clearly stated in more than one place:
The first post of the thread says:
- x86 does not currently support Nv2 (requires RouterBOARD hardware)
- x86 may be supported in the future
I didn't download the packages. I just implement what my superiors tell me to do. Their instruction was to upgrade the network of close to 1100 routers to ver 5.00 The only reason why I came here was to enquire about the user manager. Anyway, it doesn't make sense to me that Nv2 works well on x86 under 4.17, but won't work on 5.00
The x86 DOWNLOAD page of mikrotik officially states:
- x86 does not currently support Nv2 (requires RouterBOARD hardware)
- x86 may be supported in the future
On RB433AH 'Overall Stats', 'RX Stats' and 'TX Stats' tabs are filled with zeros, not with stats data.*) winbox - added more detailed /interface ethernet stats;
The same problem but fw upgrade passed successfully from terminal window.I have problem with CPU load 100% after upgrading from 4.16 to 5.0 all the time. Before it was never 100% all the time.
I can't upgrade firmware from 2.28 to 2.33. I give upgrade command and get message that it is upgraded sucessfully, and to reboot it. After reboot I see that it is still on 2.28 version firmware, and that available version for upgrade is 2.33.
Also, in winbox new version when I open interfaces windows and ppp window I don't see anything. After some time I get message in this window ERROR: timeout
It is RB750 router. Can anyone help me with this problems.
Regards,
Sasa
Sometimes you have to do the firmware update 4 or 5 times to get it to work. That's how I got mine to work on RB750... just keep trying it, and rebooting... and it will finally work.
Version 4 has one particular issue I'm wondering about; IPSEC SA's usually have to be flushed if a tunnel is interrupted between a RouterOS device and a non-RouterOS device. Has this been corrected in Version 5?
Hi,
There is a still issue with BGP (session only on ethernet/tunnels) and WLAN (no bgp peers), where there switch and someone is present on WLAN and then disconnect.
After this, routing gets 100% CPU usage on RB433. Only way is reboot router.
Support: Ticket#2011021266000179 (sent last supout.rif from 5.0)
I wondering when it's will be fixed, or I have to buy standalone access point.
I think that if the WLAN interface will be always up, even there is no client then should be working without crashing routing process.
Do you have the same issue on just one or few RouterBOARDs?The same problem but fw upgrade passed successfully from terminal window.I have problem with CPU load 100% after upgrading from 4.16 to 5.0 all the time. Before it was never 100% all the time.
I can't upgrade firmware from 2.28 to 2.33. I give upgrade command and get message that it is upgraded sucessfully, and to reboot it. After reboot I see that it is still on 2.28 version firmware, and that available version for upgrade is 2.33.
Also, in winbox new version when I open interfaces windows and ppp window I don't see anything. After some time I get message in this window ERROR: timeout
It is RB750 router. Can anyone help me with this problems.
Regards,
Sasa
Thank you very much for the problem. We will test it on Monday.Hi,On RB433AH 'Overall Stats', 'RX Stats' and 'TX Stats' tabs are filled with zeros, not with stats data.*) winbox - added more detailed /interface ethernet stats;
Bug or I missed something?
Currently I'm not able to test this on other RBs.
Regards,
Seems to be good workaround but I don't trust it.drixster, will putting the AP in a bridge and run your stuff off bridge?
Bridge is always up.
/system license print
Thank you, I tried DPD and it seems to work.There is one option added,
*) ipsec - added command kill-connections under remote-peers;
DPD allows to overcome the problem too, but it should be supported by both ends.
Low value for lifetime on both ends could be used too.
I have been using NV2 protocol on x86 boards since 5.0RC5... I admit, the RC11 wasn't passing traffic on a WDS bridged link (my post here) but a downgrade to RC10 solved my problem.Not a joke, it's been noted for at least a few days that it doesn't work with x86... although if you look back, at least 1 person got nv2 to work with x86 anyway.... Well I really hope this is an April fools joke...you are wrong, Nv2 is very much architecture dependent, there are many technical reasons which cause difficulties.
heres the outputHave you updated the licence to the new 8-digit format?
What is the output ofCode: Select all/system license print
Well that says you can only upgrade to v4heres the outputHave you updated the licence to the new 8-digit format?
What is the output ofCode: Select all/system license print
[myron@Kamote] /system license> print
software-id: "9CR1-V147"
upgradable-to: v4.x
nlevel: 5
features:
Then upgrade to 4.17 and you will from then on be able to upgrade to 5.0
Looks like my RB133c problem. Ethernet link up but RB considered it down.yesterday our routeros v5.0 sudenly crached, it was on, lan was blinking but no ping, no winbox search, so when i restart it it works again.
downgrade to 4.17 again
please create supout.rif file on that router and send it to support@mikrotik.comI find out what is causing 100% CPU load all the time. It is ppp process when I look in profile menu it is consuming from 75-93% CPU. But there is anything which can so much load CPU (I have one pptp-out and one ovpn-out) connection.
Have anyone any idea about this?
[admin@MikroTik] > interface bridge port remove [find]
[admin@MikroTik] > interface bridge remove [find]
[admin@MikroTik] > interface ethernet switch set switch1 switch-all-ports=no
mirror-source does not belong to switch
[admin@MikroTik] >
It is shown because of the bug. It will be fixed.what does it mean with "mirror-source does not belong to switch" ?
Nv2 protocol can be enabled on x86 and will allow this setting in the future. We have had less than ten reports of the problems below, so we consider that it may be an issue specific to certain x86 hardware configurations -- but we can not guarantee that you will not encounter this problem. If you must use an x86 device with Nv2 protocol, it is suggested that you test and monitor a device before deploying it broadly through your network.- x86 does not currently support Nv2 (requires RouterBOARD hardware)
x86 may be supported in the future
And THAT is how you say a feature that was in testing has been removed from releaseI would like to clear up some confusion about x86 and Nv2 protocol issues that are mentioned on the download pageNv2 protocol can be enabled on x86 and will allow this setting in the future. We have had less than ten reports of the problems below, so we consider that it may be an issue specific to certain x86 hardware configurations -- but we can not guarantee that you will not encounter this problem. If you must use an x86 device with Nv2 protocol, it is suggested that you test and monitor a device before deploying it broadly through your network.- x86 does not currently support Nv2 (requires RouterBOARD hardware)
x86 may be supported in the future
These problems may affect clients and APs, but as far as we can determine the issue only becomes a problem when seen on an x86 AP.
We have seen 2 issues with x86 boxes that are using Nv2 wireless protocol on v4.x (with wireless-nv2 package installed), v5.0rcX and v5.X:
1) You will experience unstable performance/disconnections when using x86 router in Nv2 wireless protocol network.
2) We have received reports from clients that their x86 routers hangs/freezes when they use Nv2 wireless protocols on x86 router.
About the first issue - it is hardware related and currently we haven't found any workaround for that.
About the second issue, we were unable to reproduce this problem, that is why we haven't found a solution for that.
Is how you insult your clients. I doubt anyone outside of MT understands the in's and out's of your NV2 protocol so when you pull NV2 support from x86 the rest of the world is going "what?" a CPU is a CPU.you are wrong, Nv2 is very much architecture dependent, there are many technical reasons which cause difficulties.
I've tested 5.0 on a RB1100 with a full BGP feed. CPU is at about 80-100% usage constantly. MT says it has to do with new routecalculation and that it's 'normal'. I'm going to run with three (maybe four) full feeds and will be buying x86 hardware for that.So is the BGP memory leak fixed?
I'm still running 4.13 that came installed on our RB1100, but it's leaking memory running BGP. We're taking full routes from one peer and just default route from another. So about 350k prefixes. Memory use seems to creep up... And now I'm wondering should I just go to 4.17 or 5.0 here.
(as you can see, we started running BGP mid-week 12)
This is correct, that in v5.0 you will see higher CPU usages with BGP. It is due to more frequent nexthop calculations which is made to address several problems where nexthops were not updated correctly. CPU usage depends on how frequently peer receives updates (more frequent updates will cause higher cpu usage).I've tested 5.0 on a RB1100 with a full BGP feed. CPU is at about 80-100% usage constantly. MT says it has to do with new routecalculation and that it's 'normal'. I'm going to run with three (maybe four) full feeds and will be buying x86 hardware for that.So is the BGP memory leak fixed?
I'm still running 4.13 that came installed on our RB1100, but it's leaking memory running BGP. We're taking full routes from one peer and just default route from another. So about 350k prefixes. Memory use seems to creep up... And now I'm wondering should I just go to 4.17 or 5.0 here.
(as you can see, we started running BGP mid-week 12)
IMHO, the RB1100 with 5.0 is not suitable for full BGP feeds.
Read again my post. We haven't removed anything, you can still choose Nv2 protocol on x86 router.And THAT is how you say a feature that was in testing has been removed from release
Please write to support with that report and include the support output files, we will try to look more detailed.Uldis That will tell about the report nv2 on distances more than 40 km ?
Whether the help in testing is necessary?
So is the 1100AH enough hardware to throw at a full feed BGP router? I do not want to go to x86 hardware, BGP is the reason I bought a RB1100 in the first place... And if I can get the CPU to handle it, is the memory leak still there?This is correct, that in v5.0 you will see higher CPU usages with BGP. It is due to more frequent nexthop calculations which is made to address several problems where nexthops were not updated correctly. CPU usage depends on how frequently peer receives updates (more frequent updates will cause higher cpu usage).
We are working to improve performance, but until then users who experience too high CPU usage will require faster hardware.
We are not able to repeat any memory leaks in v5.So is the 1100AH enough hardware to throw at a full feed BGP router? I do not want to go to x86 hardware, BGP is the reason I bought a RB1100 in the first place... And if I can get the CPU to handle it, is the memory leak still there?This is correct, that in v5.0 you will see higher CPU usages with BGP. It is due to more frequent nexthop calculations which is made to address several problems where nexthops were not updated correctly. CPU usage depends on how frequently peer receives updates (more frequent updates will cause higher cpu usage).
We are working to improve performance, but until then users who experience too high CPU usage will require faster hardware.
VirtualAP feature isn't made for the Nv2 wireless protocol. For the rest wireless protocols it should be working.Virtual Ap seem to still be not working. Is this correct?
seems like you're using certificate for that user - that way, you cannot connect via ssh with password; use your certificate. or simply remove all ssh keysi just upgrade RB433 from 4.11 to v5 and canot ssh to router .
it accept connection but dont accept password.
Same password works over winbox on same router.
/user ssh-keys remove [find]
working on this.Hello,
I have noticed that in webfig ->IP->Firewall->NAT list the Src. Port and Dst.Port columns are empty while in the same time the webfig detailed view of the rule shows correctly.
This is not a major issue, but it is hard to identify the right rule that needs to be changed.
Thank you in advance.
Regards,
Z
This will cause problem but will open new topic. Thanks Aldus.VirtualAP feature isn't made for the Nv2 wireless protocol. For the rest wireless protocols it should be working.Virtual Ap seem to still be not working. Is this correct?
Unfortunately, not on mineworking on this.I have noticed that in webfig ->IP->Firewall->NAT list the Src. Port and Dst.Port columns are empty while in the same time the webfig detailed view of the rule shows correctly.
This is not a major issue, but it is hard to identify the right rule that needs to be changed.
Did you set any src or dst address at all? In your case both config windows don't show them.
Unfortunately, not on mine
Could you please help me what I am doing wrong.
Thank you in advance.
Regards,
Z
[offtop]where did you see that? %)[/offtop]you are using a very old version of internet explorer
Here you are, MT guru's always know more than us![offtop]where did you see that? %)[/offtop]you are using a very old version of internet explorer
I had the same problem.Same problem. RB433 4.16 to 5.0 Upgrade Fail (Disconnect)I upgraded an RB running 4.16 to 5.0 that had the NV2 package installed and when the router came back up it didn't have any wireless package enabled and lost all the wireless settings. Is there any way around this issue?
I lucked out in this case because it was the local side that I upgraded, but it could have resulted in a frustrating truckroll.
NV2 package has been disabled
I have problems showing the src and dst ports and not addresses... the addresses are shown, but not the ports.Did you set any src or dst address at all? In your case both config windows don't show them.Unfortunately, not on mine
Could you please help me what I am doing wrong.
Double click on the rule to see what is actually set. You'll find much more settings not available in the default colums...
I use Firefox on Linux and Chrome on Windows environment with same resultAlso try another browser, you are using a very old version of internet explorer, it has thousands of bugs.
yep, that 'strange' command gives strange result. correct one is "=interface=ether1", not "?interface=ether1"Problem with API in 5.0??
/interface/monitor-traffic
?interface=ether1
=once=
result:
!re=rx-packets-per-second=49=rx-drops-per-second=0=rx-errors-per-second=0=tx-packets-per-second=56=tx-drops-per-second=0=tx-errors-per-second=0
where is tx/rx bites/s ??
Indeed the direct download failed... but the torrent download went perfect (and I'll be in the seed for a while for ALL)Torrent files seems to be "borked" at least for mipsbe architecture
I don't know, but we are with 5.7 now. 5.0 is old and had lots of issues. Hence 5.7 now....Thanks for the reply! Interesting... Perhaps I will revisit after some sleep then...
So what is the current status? Does it work or not? Some people do prefer to use their own boards Atom etc.!What's new in 5.0 (2011-Mar-31 11:33):
...
On MT official site is written:
x86 does not currently support Nv2 (requires RouterBOARD hardware)
x86 may be supported in the future
but i tried in on x86 and it works
nothing has changed regarding this issue. The above statement is still valid.So what is the current status? Does it work or not? Some people do prefer to use their own boards Atom etc.!What's new in 5.0 (2011-Mar-31 11:33):
...
On MT official site is written:
x86 does not currently support Nv2 (requires RouterBOARD hardware)
x86 may be supported in the future
but i tried in on x86 and it works